After a number of unplanned distractions (involving the story behind the Archaeopteryx forgery claim, the time-honoured tradition that is April 1st, feathered tyrannosaurs, horned dinosaurs, chickens, ‘Cadborosaurus’, Eld’s deer, and intraguild predation in, and the phylogeny of, raptors), it’s time to get back on track and carry on looking at tubenose seabirds, and petrels in particular. If you need a refresher on where we got to, the previous articles are here (part I), here (part II) and here (part III).

Is an albatross (like the Wandering albatross (Diomedea exulans) shown in the middle) more like a petrel (like the Great shearwater (Puffinus gravis) shown at left) or like a storm-petrel (like the Wilson's storm-petrel (Oceanites oceanicus) shown on the right)? Images by Patrick Coin (l and r) and JJ Harrison (centre), licensed under Creative Commons Attribution-Share Alike 3.0 Unported license and Creative Commons Attribution-Share Alike 2.5 Generic license.

Petrels (by which I mean ‘true petrels’) have conventionally been regarded as a ‘family’ (termed Procellariidae) within the ‘order’ Procellariiformes (popularly termed tubenoses). But petrels aren’t the only tubenose ‘family’. Before moving on to look at petrel phylogeny and diversity in full, glorious detail, my plan here is to look at the position of petrels within tubenose phylogeny as a whole. How are petrels related to other tubenoses? And, if we know that, what might it tell us about tubenose evolution as a whole?

The ‘four families’ system breaks down

Shy albatross (Thalassarche cauta), 'super petrel' or something else entirely? Those interested in the soft-tissue reconstruction of fossil archosaurs will already be enamored with the cheek anatomy of this (and a few other) albatross species. Photo by JJ Harrison, licensed under Creative Commons Attribution-Share Alike 3.0 Unported license.

I grew up with the idea that petrels were especially closely related to the largest, most spectacular and most famous of tubenosed seabirds, the albatrosses. In fact my impression as a rather younger person was that albatrosses were ‘super petrels’, a notion mostly based on Sibley & Ahlquist’s (1990) contention that the two groups were closely related ‘subfamilies’ that only diverged about seven million years ago. Today it seems that albatrosses are not especially close to petrels proper, and indeed views on the phylogeny of tubenoses as a whole don’t support the ‘family-level’ classification used in most 20th century texts, partly because storm-petrels (hydrobatids) seem non-monophyletic (Nunn & Stanley 1998, Kennedy & Page 2002, Hackett et al. 2008).

Interested in testing correlations between body size and rates of molecular evolution, Nunn & Stanley (1998) produced what I think was the first large-scale molecular analysis of tubenoses. As just mentioned, storm-petrels were not recovered as monophyletic, with hydrobatines (one of two storm-petrel clades) being closer to the remaining lineages than was the other storm-petrel clade. The next major divergence involved albatrosses and a diving-petrel + true petrel clade. Kennedy & Page (2002) showed via a supertree approach that studies as of that time generally supported the following topology: (hydrobatine storm-petrels + (albatrosses + (oceanitine storm-petrels + true petrels))). Diving-petrels were deeply nested within true petrels. Seeing as diving-petrels were generally regarded then as a ‘family’ (Pelecanoididae), finding them nested within ‘family’ Procellariidae (true petrels) made Procellariidae non-monophyletic as well.

Images by (l to r) Bryan Harry, Sannab, Patrick Coin, Júlio Reis and James Lloyd; images licensed under Creative Commons Attribution ShareAlike 3.0 License, Creative Commons Attribution-Share Alike 3.0 Unported license, Creative Commons Attribution-Share Alike 2.5 Generic license, and Creative Commons Attribution-Share Alike 3.0 Unported license.

Penhallurick & Wink (2004) used mitochondrial cytochrome b sequence data to look both at higher-level relationships among tubenoses, and to assess the classification of taxa at the genus, species and subspecies level. Again, storm-petrels were recovered as non-monophyletic, but albatrosses now grouped with the two storm-petrel clades, being closer to hydrobatines in their favoured tree. Shock horror. Prions, diving-petrels and true petrels formed the sister-group to the storm-petrel + albatross clade, though the position of prions (conventionally included within true petrels) was fairly labile (Penhallurick & Wink 2004). In their favoured phylogeny, diving-petrels were nested within true petrels, being especially close to gadfly petrels. Ericson et al. (2006) found albatrosses to be outside a (storm-petrel + (petrel + diving-petrel)) clade.

Click to enlarge. Attributions as above but prion image by Rosemary Tully, licensed under Creative Commons Attribution-Share Alike 3.0 Unported license.

Livezey & Zusi (2007) didn’t include many taxa in their famously large analysis, and – in contrast to most other recent studies – found true petrels and albatrosses to be sister-groups, with prions, the storm-petrel Oceanites and diving-petrels to be successively more distant to this clade. It has been argued that at least some of the character codings used by Livezey & Zusi (2007) are inaccurate, with taxa being coded for characters that they ‘should’ have, rather than what they do have (Mayr 2008). Furthermore, some of the relationships they recovered (e.g., a loon-grebe clade) are worryingly ‘traditional’.

Finally (for now), Hackett et al. (2008) found the oceanitine storm-petrel Oceanites to be outside a tubenose clade that includes albatrosses, hydrobatine storm-petrels, diving-petrels and true petrels.

Tubenose phylogeny recovered by Hackett et al. (2008). Image attribution as above.

Some implications, and what do the fossils say?

What ‘consensus’ emerges from these studies, and what does it actually mean that’s actually, you know, interesting? For starters, the tidy, apparently traditional view that Procellariiformes consists of (1) Hydrobatidae (storm-petrels), (2) Pelecanoididae (diving-petrels), (3) Procellariidae (true petrels) and (4) Diomedeidae (albatrosses) does not accurately reflect phylogeny.

Firstly, ‘Hydrobatidae’ consists of two distinct clades and is not a natural group. Secondly, while diving-petrels remain monophyletic, their classification as a ‘family’ will now confuse many given that they might be nested within true petrels. As noted above, Penhallurick & Wink (2004) advocated the view that diving-petrels are within a true petrel clade that also includes gadfly petrels, and they thus treated diving-petrels as a ‘tribal-level’ clade (termed Pelecanoidini) within a ‘subfamily-level’ clade (termed Pelecanoidinae) within ‘family’ Procellariidae.

Tubenose phylogeny from Penhallurick & Wink (2004), with clades labelled. I don't expect for a moment that you'll be able to read much of the text here, but you should be ale to see that they classified all crown-tubenoses as either within Diomedeidae, or within Procellariidae. Actually, in the tree shown here, prions are incertae sedis, but they were recovered as part of Procellariidae in other trees.

Penhallurick & Wink’s (2004) phylogeny – while not exactly a ‘last word’ on the subject – would actually make the big picture of crown-tubenose phylogeny simpler than has been conventional. This is because members of the tubenose crown-group are, in their phylogeny, either members of the albatross lineage, or members of the true petrel lineage: Penhallurick & Wink’s (2004) therefore proposed that both kinds of storm-petrel should be included within Diomedeidae alongside albatrosses (which then become the ‘subfamily-level’ clade Diomedeinae); the true petrel lineage (including diving-petrels) would obviously be termed Procellariidae.

Anyway, whichever phylogenetic hypothesis we follow, it does seem generally agreed that albatrosses are not at all close to true petrels. In fact, the relatively enormous, hyper-long-winged, soaring, often mostly white albatrosses perhaps evolved from a small, dark, storm-petrel-like ancestor given that albatrosses are nested within storm-petrels according to Penhallurick & Wink (2004), and surrounded by them according to Kennedy & Page (2002) and Hackett et al. (2008).

Are there any fossils that might shed light on the ancestral conditions of albatrosses and other crown-tubenoses? Most fossil tubenoses aren’t all that different from living ones. There are, however, a few peculiar specimens, among them the comparatively tiny albatross Murunkus subitus from the Eocene of Uzbekistan, known only from its carpometacarpus. This is seemingly from a bird about a third smaller than the smallest living albatross; so, with a wingspan of perhaps 60 cm or so. However, it’s an ‘alleged’ albatross and its identification as a member of the group awaits confirmation (Mayr & Smith 2012). Other fossil albatrosses (the oldest definite record, Tydea septentrionalis, is from the Oligocene of Belgium) are much like living ones in form, size, and probably in ecology and behaviour.

Foot skeletons of the Oligocene diomedeoidid tubenose Diomedeoides brodkorbi, from Mayr (2009b). The relatively broad, but long, toe bones and their 'closely packed' appearance is obvious, as is the blunt form of the claws. The tiny hallux is a key tubenose character.

Members of an entirely extinct tubenose group are known from the Oligocene of Europe and Iran. These are the diomedeoidids. Yeah, I know, that name is horrible – too many vowels, and too similar to Diomedeidae. Anyway, based on the small size of the dorsal supracondylar process on the distal end of the humerus and other characters, it’s been inferred that diomedeoidids are outside the crown-tubenose clade (Mayr 2009a, De Pietri et al. 2010). The suggestion that this group might be stem-tubenoses explains why I’ve sometimes made a distinction in this article between Procellariiformes as generally understood, and crown-Procellariiformes. Mayr (2009a, b) suggested, on the basis of that small supracondylar process and perhaps other features, that diomedeoidids may have been flap-gliders like oceanitine storm-petrels, and not gliders or soarers. Their very long legs recall those of oceanitines, and their peculiar wide, flattened toes (with blunt, nail-like claws) do too.

Wilson's storm-petrel (Oceanites oceanicus), practicing surface-pattering. Image by Patrick Coin, licensed under the Creative Commons Attribution-Share Alike 2.5 Generic license.

I can’t pretend that we really know all that much about diomedeoidid ecology, functional morphology or behaviour. But there are indications that these possible stem-tubenoses were storm-petrel-like ‘surface patterers’, repeatedly braking to pick up prey from the sea surface, and not soaring swiftly and efficiently on stiff wings like large petrels or albatrosses. [Adjacent image of 'surface patterning' storm-petrel by Patrick Coin].  And if this is correct, and if albatrosses evolved their stiff-winged, high-aspect-ratio soaring wings from storm-petrel-like ancestors (as indicated by some of the phylogenetic results discussed above), then any ecological, behavioural and morphological similarities shared by albatrosses and true petrels must represent convergences. When you look at the general similarity between, say, giant petrels (Macronectes) and albatrosses, that seems pretty surprising.

Much as I’d like to discuss other aspects of tubenose evolution and historical biology, I need to finish this article by briefly discussing the phylogenetic structure of true petrels (Procellariidae), since that’s where we going next.

A phylogeny for petrels

Fulmarus glacialis, a member of Fulmarini. Image by T. Müller, licensed under Creative Commons Attribution-Share Alike 3.0 Unported license.

Recent molecular phylogenetic studies indicate that true petrels (Procellariidae) consist of four major clades: Pterodromini (gadfly-petrels), Procellarinii, Fulmarini (fulmars, giant petrels and kin) and Puffinini (shearwaters). The majority of analyses have found Procellarinii and Puffinini to be sister-taxa, with Fulmarini and Pterodromini representing successively more distant clades to this pairing (Bretagnolle et al. 1998, Nunn & Stanley 1998, Kennedy & Page 2002, Penhallurick & Wink 2004).

The situation is somewhat complicated by the fact, discussed above, that diving-petrels – traditionally classified within their own ‘family’ (Pelecanoididae) – form the sister-taxon to pterodromines in some studies. If this phylogeny is followed, it might be appropriate to split Procellariidae into Procellariinae (for Procellarinii, Fulmarini and Puffinini) and Pelecanoidinae (for Pterodromini and the diving-petrels). Furthermore, prions – conventionally regarded as part of Procellarinii – were found to be outside of the diving-petrel + true petrel clade in some of the topologies recovered by Penhallurick & Wink (2004).

The positions of some species are fairly labile in analyses. Consequently, while the existence of those four major clades (Pterodromini, Procellarinii, Fulmarini and Puffinini) is generally agreed upon (Bretagnolle et al. 1998, Nunn & Stanley 1998, Kennedy & Page 2002, Penhallurick & Wink 2004), their membership varies between analyses. When we visit petrels again, we’ll be looking at pterodromines.

For previous Tet Zoo articles on petrels and other tubenosed seabirds, see…

• A symbiotic relationship between sunfish and… albatrosses? Say what?
• Because the world belongs to petrels (petrels part I)
• Living the pelagic life: of oil, enemies, giant eggs and telomeres (petrels part II)
• Petrels: some form-function ‘rules’, and pattern and pigmentation (petrels part III)
• Noel W. Cusa’s brilliant seabird drawings

And for articles about other kinds of seabirds, see…

• To the Sahara in quest of dinosaurs (living and extinct) (includes discussion of gulls and terns)
• Gannets, most awesome of seabirds
• Happy 2009, from the gulls
• Fascinated by boobies
• Mysterious channels of Alca torda
• Kleptoparasitism at Westbury Manor
• Gary Kaiser’s The Inner Bird: Anatomy and Evolution
• When bivalves attack (or: bivalves vs birds, the battle continues)

Refs – -

Bretagnolle, V., Attié, C., Pasquet, E. 1998. Cytochrome-B evidence for validity and phylogenetic relationships of Pseudobulweria and Bulweria (Procellariidae). Auk 115, 188-195.

De Pietri, V. L., Berger, J.−P., Pirkenseer, C., Scherler, L. & Mayr, G. 2010. New skeleton from the early Oligocene ofGermany indicates a stem−group position of diomedeoidid birds. Acta Palaeontologica Polonica 55, 23–34.

Ericson, P. G. P., Anderson, C. L., Britton, T., Elzanowski, A., Johansson, U. S., Källersjö, M., Ohlson, J. I., Parsons, T. J., Zuccon, D. & Mayr, G. 2006. Diversification of Neoaves: integration of molecular sequence data and fossils. Biology Letters 2, 543-547

Hackett, S., Kimball, R., Reddy, S., Bowie, R., Braun, E., Braun, M., Chojnowski, J., Cox, W., Han, K., Harshman, J., Huddleston, C., Marks, B., Miglia, K., Moore, W., Sheldon, F., Steadman, D., Witt, C., & Yuri, T. (2008). A Phylogenomic Study of Birds Reveals Their Evolutionary History Science, 320 (5884), 1763-1768 DOI: 10.1126/science.1157704

Kennedy, M. & Page R. D. M. 2002. Seabird supertrees: combining partial estimates of procellariform phylogeny. Auk 119, 88-108.

Livezey, B. C. & Zusi, R. L. 2007. Higher-order phylogeny of modern birds (Theropoda, Aves: Neornithes) based on comparative anatomy. II. Analysis and discussion. Zoological Journal of the Linnean Society 149, 1-95.

Mayr, G. 2008. Avian higher- level phylogeny: well-supported clades and what we can learn from a phylogenetic analysis of 2954 morphological characters. Journal of Zoological Systematics and Evolutionary Research 46, 63-72.

- . 2009a. Paleogene Fossil Birds. Springer, Berlin.

- . 2009b. Notes on the osteology and phylogenetic affinitiesof the Oligocene Diomedeoididae (Aves, Procellariiformes). Fossil Record 12, 133–140.

- ., & Smith, T. 2012. A fossil albatross from the Early Oligocene of the North Sea Basin. The Auk 129, 87-95.

Nunn, G. B. & Stanley, S. E. 1998. Body size effects and rates of cytochrome b evolution in tube-nosed seabirds. Molecular Biology and Evolution 15, 1360-1371.

Penhallurick, J. & Wink, M. 2004. Analysis of the taxonomy and nomenclature of the Procellariiformes based on complete nucleotide sequences of the mitochondrial cytochrome b gene. Emu 104, 125-147.

Sibley, C. G. & Ahlquist, J. A. 1990. Phylogeny and Classification of Birds. New Haven: Yale University Press.